Certified into the record
The exact text we saved
This is the version exactly as it was saved, character for character. The SHA-256 hash below is a 64-character code worked out from that text. Run the same text through SHA-256 and you get the same code; change a single letter and the code comes out completely different. Each version also stores the previous version’s hash, which chains them together so no version can be swapped out quietly.
TRUVACE RECORD VERSION record: TRV-2026-3062 version: 1 kind: certified reason: Certified into the record timestamp: 2026-11-19T18:03:10.033744Z status: published lens: g_space sector: business headline: The association between different types of anthropomorphism of artificial intelligence service agents and user trust: A three-level meta-analysis dek: As artificial intelligence service agents (AISAs) become increasingly widespread, anthropomorphism is often proposed as a way to foster user trust, yet empirical findings remain mixed. Drawing on competing trust-building and trust-undermining mechanisms, we conducted a three-level meta-analysis of 211 effect sizes from 136 independent samples ( N = 46,071) to estimate the anthropomorphism–trust association and examine its boundary conditions. The average association was positive and robust to alternative depende… gain_title: A meta-analysis of 136 independent samples (46,071 people) found that users who saw AI service agents as more human-like reported higher trust in them (r = 0.462), though the link was weaker (r = 0.222) for objectively specified design cues. problem_title: (none) trace_subject: (none) gain_reading: A meta-analysis of 136 independent samples (46,071 people) found that users who saw AI service agents as more human-like reported higher trust in them (r = 0.462), though the link was weaker (r = 0.222) for objectively specified design cues. gain_evidence: It was considerably weaker for objectively specified design cues ( r = 0.222) than for perception-based measures ( r = 0.543) problem_reading: (none) problem_evidence: (none) quick_read: A three-level meta-analysis pooled 211 effect sizes from 136 independent samples covering 46,071 people. It found a positive average association between how human-like an AI service agent seems and how much users trust it (r = 0.462, 95% CI 0.420 to 0.502). The result is not uniform. The 95% prediction interval runs from -0.131 to 0.811, so in some settings human-like cues were linked to less trust. The association was much weaker for objectively specified design cues (r = 0.222) than for perception-based measures (r = 0.543), most of which captured traits such as competence and warmth. The authors suggest that human-like cues should complement, rather than replace, demonstrable competence and reliability when AI service agents are designed and deployed. limitation: The authors say construct overlap between anthropomorphism and trust-related outcomes cannot be ruled out, and no design-based cognitive anthropomorphism effect sizes were available, so the pooled association should not be read as a uniform effect of design. tag: Evidence-backed gain key_points: The analysis combined 211 effect sizes from 136 independent samples with 46,071 participants in total. | The pooled link between anthropomorphism and trust was r = 0.462, and a restricted analysis of global trust, reliance or willingness to depend gave a similar r = 0.475. | The prediction interval of -0.131 to 0.811 shows the association varied substantially across settings. | Objectively specified design cues were linked to trust far more weakly (r = 0.222) than perception-based measures (r = 0.543). | Cognitive and emotional anthropomorphism were more strongly associated with trust than physical anthropomorphism overall, but that gap was no longer significant after adjusting for how it was measured. | The association was stronger in Eastern than Western samples and in survey than experimental studies; no significant moderation appeared for gender composition or the prespecified service-context categories. rundown: The researchers drew on competing explanations: that human-like features build trust, and that they can undermine it. They tested these by pooling studies of AI service agents and estimating the anthropomorphism-trust association with a three-level model that accounts for multiple effect sizes from the same sample. How anthropomorphism was measured mattered. Perception-based measures, most of which captured adjacent trait attributions such as competence and warmth, showed a stronger link (r = 0.543) than objectively specified design cues (r = 0.222). Among design cues, physical and emotional anthropomorphism did not differ significantly, and no design-based cognitive effect sizes were available. Context also shifted the result: the association was stronger in Eastern than Western samples and in survey than experimental studies. The authors conclude that the pooled figure should not be read as a uniform effect of anthropomorphic design. sources: - peer_reviewed | Technological Forecasting and Social Change | https://doi.org/10.1016/j.techfore.2026.124959 | 2026-11-18 prev: 0000000000000000000000000000000000000000000000000000000000000000
- SHA-256 hash
- 0680ec286aa37316d230cadeb3bdd9e7feb2d996c7f8474c4225ac5b32e30118
- Previous version’s hash
- 0000000000000000000000000000000000000000000000000000000000000000
How to check it yourself, without trusting this page
Download any version from /api/record/TRV-2026-3062 and hash its canonical field yourself, for example with shasum -a 256. You should get its content_hash. Each version’s text also ends with prev:and the hash of the version before it (version 1 points to 64 zeros). If anyone had changed a single character of any version after it was saved, those hashes wouldn’t line up.
ace